…When to Think While Listening in Large Audio-Language Models (2026) Segment-Aligned Policy Optimization for Multi-Modal Reasoning (2026) RASFT: Rollout-Adaptive Supervised Fine-Tuning for Reasoning (2026) AdaptR1: Reinforcement Learning…
…latency, streaming, and controllable rollout. Existing autoregressive (AR) diffusion distillation methods have achieved strong results in the chunk-wise 4-step regime by distilling bidirectional base models into few-step AR students…
…phased teacher sampling mixes in verifier-endorsed teacher rollouts at cold-start, and a sign-consistency gate extrapolates the distillation update on tokens where the teacher agrees with the verifier-correct direction…
…Aligning Diffusion Models with Fine-Tuning Is Easier Than You Think (2026) Reward Sharpness-Aware Fine-Tuning for Diffusion Models (2026) Step-level Denoising-time Diffusion Alignment with Multiple Objectives (2026) Astrolabe…
…even as the rollout progresses. AdaState replaces this static anchor with a self-evolving one. We reserve a hidden latent slot inside the KV cache that the model denoises alongside each chunk…
…Hidir Yesiltepe , , , , , , Abstract VideoMLA reduces memory usage in video diffusion models by replacing per-head keys and values with shared low-rank content and decoupled 3D-RoPE positional keys, maintaining quality while…
…incorporating retriever-specific human guidance and by scaling model size. To facilitate learning over multi-retrieval-step trajectories , we introduce a branching-based rollout technique that improves training stability. Our work provides…
…A native inference engine enables efficient multi-agent rollout through shared visual prefix and KV cache reuse . Across V*, CountBench, the RefCOCO family, and HallusionBench, Visual Para-Thinker++ consistently outperforms single-trajectory…
…Distillation in LLM Pretraining (2026) Single-Rollout Hidden-State Dynamics for Training-Free RLVR Data Selection (2026) On the Generalization Gap in Self-Evolving Language Model Reasoning (2026) LoRi: Low-Rank Distillation…
…Shared-Prompt Flash Attention for Efficient RL Training with Large Rollouts and Long Contexts (2026) RATrain: A Resource-Aware Training Runtime for Large Language Models on Bandwidth-Constrained Heterogeneous Supercomputing Platforms (2026…